全球大陆弧玄武岩（CAB）的特征——与岛弧玄武岩（IAB）和弧后玄武岩（BAB）的对比

活动陆缘玄武岩（CAB）主要出现在安第斯,早先大多认为CAB 大体相当于IAB,在许多判别图中, CAB 基本上与IAB 在一起, 只是LILE 富集和HFSE 亏损的程度略有差别而已。本文的研究表明,尽管CAB 富集LILE,但是,Nb-Ta 亏损不是非常明显。在最新的采用大数据方法厘定的判别图中,CAB 不同于IAB,不是落在IAB 区,而是位于IAB 与OIB 的过渡区域。CAB 中保留了很多OIB 的印记,说明板块俯冲时伴随有来自地幔深部的岩浆的加入,故CAB 中保留了OIB 的印记。我们的研究还发现, 在安第斯弧后盆地玄武岩具有OIB 的印记,而非MORB 的特征。本文按照大数据的思路,讨论了CAB 的特征,并与IAB、BAB 进行了初步的对比。研究表明,CAB 明显不同于IAB,CAB 兼具IAB 和OIB 的特征,但对其深度的认识还有待于今后进一步的研究。     

三维成矿定量预测系统设计与应用实例研究

隐伏矿体三维成矿定量预测是以多元地学大数据为基础的矿产预测新技术与方法。本文基于隐伏矿体三维成矿定量预测的实际需要及相关方法步骤,采用集成二次开发的方式,设计实现了一套可在三维环境下基于大数据开展定量化成矿预测工作的软件系统。本文阐述了系统的总体架构及开发方式,并对系统中各个功能模块的设计及实现过程进行详细阐述。系统融合了当前三维成矿定量预测研究的最新方法及成果,内含数据库管理、三维地球物理正演、三维空间分析以及三维预测评价等功能模块,能够对多元地学大数据进行集成和分析预测。为了验证系统的适用性和有效性,该系统被应用于长江中下游成矿带钟姑矿田三维成矿定量预测研究,相关成果表明系统的建立不但能够深化和发展三维成矿预测理论,也为新时期基于大数据的隐伏矿体找矿勘探工作提供了新方法及有力工具。     

后弧（rear arc）岩浆作用的特征及其意义

后弧岩浆作用（rear arc magmatism）是一个新的术语,国内文献大多认为与弧后（back arc）相当,也译为弧后。实际上rear arc 不同于back arc,前者仍然属于弧的范围,而后者已不属于弧结构。目前,对后弧岩浆作用的研究还十分有限, 原因一是 rear arc 出露较少,二是 rear arc 的鉴别标志不清楚。本文尝试对后弧玄武岩（rear arc basalt, RAB）作一个简单的介绍, 并采用对大量数据进行分析比较的方法与典型的岛弧玄武岩（IAB）和弧后盆地玄武岩（BAB）作一个对比。研究表明,后弧玄武岩主要由中-高 K 钙碱性和钾玄岩系列组成, 与典型的 IAB 和 BAB 相比, RAB 富集 Na₂O、K₂O、P₂O₅ ,贫CaO。后弧岩浆作用的微量元素具有典型的弧岩浆岩的特点,但LILE 及HFSE 比典型的岛弧岩浆的含量更富集,LREE 明显高于岛弧岩浆岩。 与岛弧岩浆相似,后弧岩浆同样具有明显的Nb-Ta 负异常。研究表明,上述3 类玄武岩很难区分开。但是,BAB 和RAB之间还是有一些不同的,如Sc/Nb-Ba/Y、Cu/P₂O₅-Y/Zr、Sc/Nb-Sr/Y 以及F₂O₃ /Zr-Y/Zr 等判别图。本文作者指出,后弧岩浆作用的提出完善了弧结构：一个完整的弧,从海沟向弧的方向,随着板块的俯冲作用,岩浆源区深度增加,地壳混染程度增加,依次出现前弧、弧和后弧岩浆作用, 至弧的后部,洋壳拉张,出现弧后盆地。前弧以玻安岩为代表,弧主要是IAB,后弧为碱性玄武岩,弧后则为MORB（+IAB 的印记）。显然,后弧岩浆作用的提出,对古造山带岛弧结构的恢复、古俯冲方向的确定是有积极意义的。     

Coastal risk adaptation: the potential role of accessible geospatial Big Data

Increasing numbers of people are living in and using coastal areas. Combined with the presence of pervasive coastal threats, such as flooding and erosion, this is having widespread impacts on coastal populations, infrastructure and ecosystems. For the right adaptive strategies to be adopted, and planning decisions to be made, rigorous evaluation of the available options is required. This evaluation hinges on the availability and use of suitable datasets. For knowledge to be derived from coastal datasets, such data needs to be combined and analysed in an effective manner. This paper reviews a wide range of literature relating to data-driven approaches to coastal risk evaluation, revealing how limitations have been imposed on many of these methods, due to restrictions in computing power and access to data. The rapidly emerging field of ‘Big Data’ can help overcome many of these hurdles. ‘Big Data’ involves powerful computer infrastructures, enabling storage, processing and real-time analysis of large volumes and varieties of data, in a fast and reliable manner. Through consideration of examples of how ‘Big Data’ technologies are being applied to fields related to coastal risk, it becomes apparent that geospatial Big Data solutions hold clear potential to improve the process of risk based decision making on the coast. ‘Big Data’ does not provide a stand-alone solution to the issues and gaps outlined in this paper, yet these technological methods hold the potential to optimise data-driven approaches, enabling robust risk profiles to be generated for coastal regions.     

论"大三峡旅游经济圈"构建中的鄂西与渝东区域整合

该文论述构建“大三峡旅游经济团”的关键是实现鄂西与渝东的区域整合。鄂西与渝东在地缘和旅游开发上关系密切,但隶属不同的省级行政区,在新的形势下很有必要进行一体化开发。同时阐述区域整合和旅游一体化开发的意义,提出构建“大三峡旅游经济圈”的若干对策。     

The 'Grasses' of the Big Scrub District,North-eastern New South Wales: Their recent history,spatial distribution and origins

A feature of the pre-European landscape of the sub-tropical Richmond River district of north-eastern New South Wales was a large expanse of rainforest known as the Big Scrub. In and around the Big Scrub were small patches of grassland and grassy open-forest, known locally as 'grasses'. These were often given individual names, which indicated their importance in the early timber-based economy of this generally grassless district for camping and depasturing working stock. Historical records enable a reconstruction of the distribution of 56 named 'grasses', and also allow some inferences to be made about their botany and ecology. The 'grasses' appear to be natural features of the landscape, mainly relict areas following invasion of the late Pleistocene open-forest vegetation by rainforest, following sea level rise, during the Holocene. A toponymic study of the use of the term 'grass' in the Richmond River district is also included.     

温棚樱桃生产丰歉年的气候分析

通过对大棚内外大樱桃萌芽到成熟期间空气温度和湿度的观测 ,结合其物候期 ,分析了温湿度与樱桃生长发育及产量的关系。结果表明 ,棚内温度条件明显优越于露地 ,但温棚内温湿度调控是否得当是影响大棚樱桃产量的关键。     

World Heritage in danger: Big data and remote sensing can help protect sites in conflict zones

World Heritage sites provide a key mechanism for protecting areas of universal importance. However, fifty-four UNESCO sites are currently listed as “In Danger”, with 40% of these located in the Middle East. Since 2010 alone, thirty new sites were identified as under risk globally. We combined big-data and remote sensing to examine whether they can effectively be used to identify danger to World Heritage in near real-time. We found that armed-conflicts substantially threaten both natural- and cultural-heritage listed sites. Other major risks include poor management and development (globally), poaching (Africa mostly) and deforestation (tropics), yet conflict is the most prominent threat. We show that news-mining of big-data on conflicts and remote sensing of nights-lights enabled us to identify conflict afflicted areas in near real-time. These findings provide a crucial avenue for developing a global transparent early-warning system before irreversible damage to world heritage takes place.     

大数据与数学地球科学研究进展——大数据与数学地球科学专题代序

大数据与数学地球科学的核心应用技术包括高维数据降维、图像数据处理、无限数据流挖掘、机器学习、关联规则算法与推荐系统算法等。人工智能地质学,包括大数据-智能矿床成因模型与找矿模型的构建,是具有重要价值的研究方向。高维数据降维旨在从初始高维特征集合中选出低维特征集合,有效地消除无关和冗余特征,增强学习结果的易理解性。哈希算法、聚类分析、主成分分析等是较常用的数学降维工具。机器学习是人工智能的核心,是使计算机具有智能的根本途径。机器学习与人工智能各种基础问题的统一性观点正在形成。深度学习的训练模型往往需要海量数据作为支撑,因此迁移学习方法日益受到重视。图像模式识别是大数据挖掘的重要技术。网络中的社区结构识别对理解整个网络的结构和功能有重要价值,可帮助分析、预测网络各元素间的交互关系。沉浸式虚拟现实技术是实现大数据可视化的重要方向,对具有多元、异构、时空性、非线性、多尺度地质矿产勘查数据的展示要求有特别的价值。引入VR技术进行矿产地质大数据的可视化,可实现大数据时代矿产勘查数据的新认知。无限数据流在地质、地球化学、地球物理监测中大量存在,甚至可以持续自动产生。对数据流数据的计算包括对点查询、范围查询、内积查询、分位数计算、频繁项计算等。关联规则和推荐系统算法是大数据挖掘中的重要算法,其应用范围越来越广泛。贝叶斯原理在大数据时代有独特的价值,贝叶斯网络是成因建模的一个革命性工具。智能地质学研究刚刚起步,构建大数据-智能矿床成因模型与找矿模型是智能地质学研究的重要内容。矿床模型研究方式的变革,将出现于互联网、云计算技术环境下全球各地的矿床研究团队的共同参与。